Research map / Detection and segmentation / CNN features
AdaMixer: A Fast-Converging Query-Based Object Detector
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Enable query-adaptive sampling locations and query-adaptive content mixing (ablation) | AdaMixer with ResNet-50 · MS COCO minival, 1× | AP: 7.0 AP | Helped |
| Add adaptive channel mixing before adaptive spatial mixing (ablation) | AdaMixer adaptive mixer · MS COCO minival | AP | Helped |
| Use only adaptive channel mixing (ablation) | AdaMixer adaptive mixer · MS COCO minival | AP | Hurt |
| Use only adaptive spatial mixing (ablation) | AdaMixer adaptive mixer · MS COCO minival | AP | Hurt |
| Reverse the mixing order to spatial mixing before channel mixing (ablation) | AdaMixer adaptive mixer · MS COCO minival | AP | Hurt |
| Add extra pyramid networks (ablation) | AdaMixer · MS COCO minival | AP | Hurt |
| Vary the number of input sampling points per group (ablation) | AdaMixer · MS COCO minival | AP | Mixed |
| Increase spatial mixing output patterns beyond 128 per group (ablation) | AdaMixer · MS COCO minival | AP | Hurt |
| Add positional information and IoF bias to query self-attention (ablation) | AdaMixer · MS COCO minival | AP | Helped |
| Add IoF bias to query self-attention (ablation) | AdaMixer · MS COCO minival | AP | Helped |
| Vary the mixer group count (ablation) | AdaMixer · MS COCO minival | FLOPs | Mixed |
| Use a single feature map instead of adaptive 3D multi-scale sampling (ablation) | AdaMixer · MS COCO minival | AP | Hurt |
Closest papers by meaning
- DECO: Unleashing the Potential of ConvNets for Query-based Detection and Segmentation (ICLR 2025 · similarity 0.86)
- EASE-DETR: Easing the Competition among Object Queries (CVPR 2024 · similarity 0.86)
- CQ-DINO: Mitigating Gradient Dilution via Category Queries for Vast Vocabulary Object Detection (NeurIPS 2025 · similarity 0.85)
- DetKDS: Knowledge Distillation Search for Object Detectors (ICML 2024 · similarity 0.85)
- Order Matters: Competition-Guided Query Ordering for RNN-Based Object Detection (NeurIPS 2026 · similarity 0.85)
- MI-DETR: An Object Detection Model with Multi-time Inquiries Mechanism (CVPR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.